Absolute Reactivity Calibration of Accelerator-Driven Systems after RACE-T Experiments

The RACE-T experiments that were held in november 2005 in the ENEA-Casaccia research center near Rome allowed us to improve our knowledge of the experimental techniques for absolute reactivity calibration at either startup or shutdown phases of accelerator-driven systems. Various experimental techniques for assessing a subcritical level were inter-compared through three different subcritical configurations SC0, SC2 and SC3, about -0.5, -3 and -6 dollars, respectively. The area-ratio method based of the use of a pulsed neutron source appears as the most performing. When the reactivity estimate is expressed in dollar unit, the uncertainties obtained with the area-ratio method were less than 1% for any subcritical configuration. The sensitivity to measurement location was about slightly more than 1% and always less than 4%. Finally, it is noteworthy that the source jerk technique using a transient caused by the pulsed neutron source shutdown provides results in good agreement with those obtained from the area-ratio technique. (authors)